library(newsflash)
library(ggalt) # github version
library(hrbrmisc) # github only
library(tidyverse)
library(ggplot2)
library(scales)
library(lubridate)
library(stringr)
wh_list <- read.csv("72attacks.csv", stringsAsFactors=F)
wh_list$keyword <- wh_list$Location
wh_list$keyword <- gsub(",.*", "", wh_list$keyword)
wh_list$keyword <- gsub("\\(.*", "", wh_list$keyword)
wh_list$keyword <- str_to_title(wh_list$keyword)
wh_list$keyword2 <- gsub(".*,", "", wh_list$Location)
wh_list$keyword2 <- str_to_title(wh_list$keyword2)
wh_list$month <- gsub(", .*", "", wh_list$Date)
wh_list$year <- gsub(".*,", "", wh_list$Date)
wh_list$start <- paste0(wh_list$month, " 1, ", wh_list$year)
wh_list$start <- mdy(wh_list$start)
wh_list$end <- wh_list$start+30
wh_list$ID<-seq.int(nrow(wh_list))
stations <- list_networks(widget=FALSE)
## # A tibble: 53 × 3
## keyword network
## <chr> <chr>
## 1 NATIONAL All National Networks
## 2 ALJAZAM Aljazeera America
## 3 BLOOMBERG Bloomberg
## 4 CNBC CNBC
## 5 CNN CNN
## 6 FBC FOX Business
## 7 FOXNEWSW FOX News
## 8 MSNBC MSNBC
## 9 INTERNATIONAL All International Networks
## 10 BBCNEWSSEG BBC News
## 11 AFFNETALL All Affiliate Networks
## 12 AFFNET_ABC ABC Affiliate Stations
## 13 AFFNET_CBS CBS Affiliate Stations
## 14 AFFNET_FOX FOX Affiliate Stations
## 15 AFFNET_MYTV MYTV Affiliate Stations
## 16 AFFNET_NBC NBC Affiliate Stations
## 17 AFFNET_PBS PBS Affiliate Stations
## 18 AFFMARKALL All Affiliate Markets
## 19 AFFMARKET_Boston Boston Affiliate Stations
## 20 AFFMARKET_Cedar Rapids Cedar Rapids Affiliate Stations
## 21 AFFMARKET_Charlotte Charlotte Affiliate Stations
## 22 AFFMARKET_Cincinnati Cincinnati Affiliate Stations
## 23 AFFMARKET_Cleveland Cleveland Affiliate Stations
## 24 AFFMARKET_Colorado Springs Colorado Springs Affiliate Stations
## 25 AFFMARKET_Columbia Columbia Affiliate Stations
## 26 AFFMARKET_Dakota Dunes SD Dakota Dunes SD Affiliate Stations
## 27 AFFMARKET_Daytona Beach Daytona Beach Affiliate Stations
## 28 AFFMARKET_Denver Denver Affiliate Stations
## 29 AFFMARKET_Des Moines Des Moines Affiliate Stations
## 30 AFFMARKET_Durham Durham Affiliate Stations
## 31 AFFMARKET_Goldsboro Goldsboro Affiliate Stations
## 32 AFFMARKET_Greenville Greenville Affiliate Stations
## 33 AFFMARKET_Hampton Hampton Affiliate Stations
## 34 AFFMARKET_Las Vegas Las Vegas Affiliate Stations
## 35 AFFMARKET_Lynchburg Lynchburg Affiliate Stations
## 36 AFFMARKET_Miami Miami Affiliate Stations
## 37 AFFMARKET_Newport KY Newport KY Affiliate Stations
## 38 AFFMARKET_Norfolk Norfolk Affiliate Stations
## 39 AFFMARKET_Orlando Orlando Affiliate Stations
## 40 AFFMARKET_Philadelphia Philadelphia Affiliate Stations
## 41 AFFMARKET_Portsmouth Portsmouth Affiliate Stations
## 42 AFFMARKET_Pueblo Pueblo Affiliate Stations
## 43 AFFMARKET_Raleigh Raleigh Affiliate Stations
## 44 AFFMARKET_Reno Reno Affiliate Stations
## 45 AFFMARKET_Roanoke Roanoke Affiliate Stations
## 46 AFFMARKET_San Francisco San Francisco Affiliate Stations
## 47 AFFMARKET_Shaker Heights Shaker Heights Affiliate Stations
## 48 AFFMARKET_Sioux City Sioux City Affiliate Stations
## 49 AFFMARKET_St. Petersburg St. Petersburg Affiliate Stations
## 50 AFFMARKET_Tampa Tampa Affiliate Stations
## 51 AFFMARKET_Virginia Beach Virginia Beach Affiliate Stations
## 52 AFFMARKET_Washington DC Washington DC Affiliate Stations
## 53 AFFMARKET_Waterloo Waterloo Affiliate Stations
## # ... with 1 more variables: date_range <chr>
colnames(stations) <- c("keyword", "station", "dates")
stations$dates <- NULL
Each incident
for (i in 1:nrow(wh_list)) {
starts <- wh_list$start[i]
ends <- wh_list$end[i]
queries <- map2(starts, ends, function(x, y) {
query_tv(wh_list$keyword[i], "attack", timespan="custom", start_date=x, end_date=y)
})
queries2 <- map2(starts, ends, function(x, y) {
query_tv(wh_list$keyword2[i], "attack", timespan="custom", start_date=x, end_date=y)
})
if (!is.null(queries[[1]])) {
queries_timeline <- map_df(queries, "timeline")
queries_df <- queries_timeline %>%
group_by(station) %>%
summarize(count=sum(value))
stat_m<- left_join(stations, queries_df)
stat_m <- subset(stat_m, !is.na(count))
stat_m$ID <- i
stat_m$total <- sum(stat_m$count)
if (i == 1) {
stations_m <- stat_m
} else {
stations_m <- rbind(stations_m, stat_m)
}
count(queries_timeline, station, wt=value, sort=TRUE) %>%
mutate(pct=n/sum(n), pct_lab=sprintf("%s (%s)", scales::comma(n), scales::percent(pct)),
station=factor(station, levels=rev(station))) -> timeline_df
gg <- ggplot(timeline_df, aes(y=station, x=pct))
gg <- gg + geom_lollipop(point.colour="steelblue", point.size=3, horizontal=TRUE)
gg <- gg + geom_text(label=timeline_df$pct_lab, hjust=-.3)
gg <- gg + scale_x_continuous(labels=percent,
breaks=seq(0, 1, by=0.2), limits=c(0, 1))
gg <- gg + labs(x=NULL, y=NULL,
title=paste0("Percent of all sentences on TV channels mentioning ", wh_list$keyword[i]),
subtitle=paste0("Search parameters: Primary [", wh_list$keyword[i], "]; Context: [attack]; n=", sum(stat_m$total)),
caption="Source: Internet Archive Television News Archive via #rstats newsflash package")
gg <- gg + theme_minimal(base_family="Arial Narrow")
gg <- gg + theme(panel.grid.major.y=element_blank())
gg <- gg + theme(panel.grid.minor=element_blank())
gg <- gg + theme(axis.line.y=element_line(color="#2b2b2b", size=0.15))
gg <- gg + theme(axis.text.y=element_text(margin=margin(r=-5, l=0)))
gg <- gg + theme(plot.margin=unit(rep(30, 4), "pt"))
gg <- gg + theme(plot.title=element_text(face="bold"))
gg <- gg + theme(plot.subtitle=element_text(margin=margin(b=10)))
gg <- gg + theme(plot.caption=element_text(size=8, margin=margin(t=10)))
print(gg)
} else if ((is.null(queries[[1]]) & !is.null(queries2[[1]]))) {
queries_timeline <- map_df(queries2, "timeline")
queries_df <- queries_timeline %>%
group_by(station) %>%
summarize(count=sum(value))
stat_m<- left_join(stations, queries_df)
stat_m <- subset(stat_m, !is.na(count))
stat_m$ID <- i
stat_m$total <- sum(stat_m$count)
if (i == 1) {
stations_m <- stat_m
} else {
stations_m <- rbind(stations_m, stat_m)
}
count(queries_timeline, station, wt=value, sort=TRUE) %>%
mutate(pct=n/sum(n), pct_lab=sprintf("%s (%s)", scales::comma(n), scales::percent(pct)),
station=factor(station, levels=rev(station))) -> timeline_df
gg <- ggplot(timeline_df, aes(y=station, x=pct))
gg <- gg + geom_lollipop(point.colour="steelblue", point.size=3, horizontal=TRUE)
gg <- gg + geom_text(label=timeline_df$pct_lab, hjust=-.3)
gg <- gg + scale_x_continuous(labels=percent,
breaks=seq(0, 1, by=0.2), limits=c(0, 1))
gg <- gg + labs(x=NULL, y=NULL,
title=paste0("Percent of all sentences on TV channels mentioning ", wh_list$keyword[i]),
subtitle=paste0("Search parameters: Primary [", wh_list$keyword[i], "]; Context: [attack]; n=", sum(stat_m$total)),
caption="Source: Internet Archive Television News Archive via #rstats newsflash package")
gg <- gg + theme_minimal(base_family="Arial Narrow")
gg <- gg + theme(panel.grid.major.y=element_blank())
gg <- gg + theme(panel.grid.minor=element_blank())
gg <- gg + theme(axis.line.y=element_line(color="#2b2b2b", size=0.15))
gg <- gg + theme(axis.text.y=element_text(margin=margin(r=-5, l=0)))
gg <- gg + theme(plot.margin=unit(rep(30, 4), "pt"))
gg <- gg + theme(plot.title=element_text(face="bold"))
gg <- gg + theme(plot.subtitle=element_text(margin=margin(b=10)))
gg <- gg + theme(plot.caption=element_text(size=8, margin=margin(t=10)))
print(gg)
} else {
stat_temp <- data.frame(keyword="", station="", count=0, ID=i, total=0)
stations_m <- rbind(stations_m, stat_temp)
}
}
Total incidents
wh_list_narrow <- select(wh_list, Location, Date, ID)
wh_list_narrow <- left_join(stations_m, wh_list_narrow, by="ID")
wh_list_narrow$loc_id <- paste0(wh_list_narrow$Location, " (", wh_list_narrow$Date, ")")
wh_list_short <- wh_list_narrow %>%
group_by(loc_id) %>%
filter(row_number()==1)
wh_list_short$keyword <- NULL
wh_list_short$station <- NULL
wh_list_short$count <- NULL
write.csv(wh_list_narrow, "wh_list_narrow.csv")
gg <- ggplot(data=wh_list_narrow, aes(x=reorder(loc_id, -ID), y=count))
gg <- gg + geom_bar(stat="identity", aes(fill=station))
gg <- gg + geom_text(data=wh_list_short, aes(y=total, label=total), size=3, hjust=-.3)
gg <- gg + coord_flip()
gg <- gg + scale_y_continuous(labels=comma, limits=c(0, 7000))
gg <- gg + labs(x=NULL, y=NULL,
title="Mentions of international attacks on TV news",
subtitle="Search parameters: Primary [Location]; Context: [attack]",
caption="Source: Internet Archive Television News Archive via #rstats newsflash package")
gg <- gg + theme_minimal(base_family="Arial Narrow")
gg <- gg + theme(panel.grid.major.y=element_blank())
gg <- gg + theme(panel.grid.minor=element_blank())
gg <- gg + theme(axis.line.y=element_line(color="#2b2b2b", size=0.15))
gg <- gg + theme(axis.text.y=element_text(margin=margin(r=-5, l=0)))
gg <- gg + theme(plot.margin=unit(rep(30, 4), "pt"))
gg <- gg + theme(plot.title=element_text(face="bold"))
gg <- gg + theme(plot.subtitle=element_text(margin=margin(b=10)))
gg <- gg + theme(plot.caption=element_text(size=8, margin=margin(t=10)))
gg
wh_list_narrow2 <- as.data.frame(wh_list_narrow)
gg <- ggplot(data=subset(wh_list_narrow, station=="FOX News"), aes(x=reorder(loc_id, -ID), y=count))
gg <- gg + geom_bar(stat="identity", aes(fill=station))
gg <- gg + geom_text(data=subset(wh_list_narrow, station=="FOX News"), aes(y=count, label=count), size=3, hjust=-.3)
gg <- gg + coord_flip()
gg <- gg + scale_y_continuous(labels=comma, limits=c(0, 7000))
gg <- gg + labs(x=NULL, y=NULL,
title="Mentions of international attacks on FOX News",
subtitle="Search parameters: Primary [Location]; Context: [attack]",
caption="Source: Internet Archive Television News Archive via #rstats newsflash package")
gg <- gg + theme_minimal(base_family="Arial Narrow")
gg <- gg + theme(panel.grid.major.y=element_blank())
gg <- gg + theme(panel.grid.minor=element_blank())
gg <- gg + theme(axis.line.y=element_line(color="#2b2b2b", size=0.15))
gg <- gg + theme(axis.text.y=element_text(margin=margin(r=-5, l=0)))
gg <- gg + theme(plot.margin=unit(rep(30, 4), "pt"))
gg <- gg + theme(plot.title=element_text(face="bold"))
gg <- gg + theme(plot.subtitle=element_text(margin=margin(b=10)))
gg <- gg + theme(plot.caption=element_text(size=8, margin=margin(t=10)))
gg
